commit 7611e2da3a8fb9bfa605ca20714e38847eb177bd Author: ModelHub XC Date: Mon Jul 13 17:21:10 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: mykor/Midm-2.0-Mini-Instruct-gguf Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..f15b49c --- /dev/null +++ b/.gitattributes @@ -0,0 +1,36 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +*.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/LICENSE.txt b/LICENSE.txt new file mode 100644 index 0000000..df5e8f4 --- /dev/null +++ b/LICENSE.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2025 KT Corporation + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/Midm-2.0-Mini-Instruct-BF16.gguf b/Midm-2.0-Mini-Instruct-BF16.gguf new file mode 100644 index 0000000..ebb1933 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-BF16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ca72406eb3f408b1af3ced70f78594dc200dfeaf9a07cab59bf60b220638d0c +size 4617053184 diff --git a/Midm-2.0-Mini-Instruct-F32.gguf b/Midm-2.0-Mini-Instruct-F32.gguf new file mode 100644 index 0000000..0c88b1f --- /dev/null +++ b/Midm-2.0-Mini-Instruct-F32.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb9bf81231ea9e0be40304befd2336134708990845283c055fac75642fb4d75f +size 9227740160 diff --git a/Midm-2.0-Mini-Instruct-IQ4_NL.gguf b/Midm-2.0-Mini-Instruct-IQ4_NL.gguf new file mode 100644 index 0000000..cbb5964 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:400b1c09569e1eefef306d9e9a3c53841c02551db361c2ffdd5960e9ea1a1105 +size 1381028224 diff --git a/Midm-2.0-Mini-Instruct-IQ4_XS.gguf b/Midm-2.0-Mini-Instruct-IQ4_XS.gguf new file mode 100644 index 0000000..8e3cf73 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48ce905fbe081b6ed843d5d134448917be3513bfc69c8d31266cb59960f7e1ce +size 1320644992 diff --git a/Midm-2.0-Mini-Instruct-Q2_K.gguf b/Midm-2.0-Mini-Instruct-Q2_K.gguf new file mode 100644 index 0000000..be96d77 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d028bf4732347b0271d838f73d19d607853fa9a1dd66e49bb59564d2b340e08 +size 975376768 diff --git a/Midm-2.0-Mini-Instruct-Q3_K_L.gguf b/Midm-2.0-Mini-Instruct-Q3_K_L.gguf new file mode 100644 index 0000000..a8908f3 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5de5eaa599e5072f8dc99d9c4a1d2c0ea1c56800a982a367eaec642ae097537 +size 1304646016 diff --git a/Midm-2.0-Mini-Instruct-Q3_K_M.gguf b/Midm-2.0-Mini-Instruct-Q3_K_M.gguf new file mode 100644 index 0000000..1263562 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96b1c928ba18712405dd0cc148b35e9b0f2bb64fea5b398e5ba31a9b0fda1042 +size 1203605888 diff --git a/Midm-2.0-Mini-Instruct-Q3_K_S.gguf b/Midm-2.0-Mini-Instruct-Q3_K_S.gguf new file mode 100644 index 0000000..48636e0 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1da4a8effb6b7acea5d6fe64b24821bdc152469b261ece68c415fc5da018b4f4 +size 1088917888 diff --git a/Midm-2.0-Mini-Instruct-Q4_0.gguf b/Midm-2.0-Mini-Instruct-Q4_0.gguf new file mode 100644 index 0000000..86c2be4 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f06814adff4847c5b46df480b03e3b896a2052918cd3cf1dc83282cd68bada1a +size 1363825024 diff --git a/Midm-2.0-Mini-Instruct-Q4_K_M.gguf b/Midm-2.0-Mini-Instruct-Q4_K_M.gguf new file mode 100644 index 0000000..b980533 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a945a20ceae202ab7a96d461fa9c3710e62deaffd6654c2106ab5e813fd8225 +size 1426272640 diff --git a/Midm-2.0-Mini-Instruct-Q4_K_S.gguf b/Midm-2.0-Mini-Instruct-Q4_K_S.gguf new file mode 100644 index 0000000..816559d --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f3f147f6538eed02f090334a708b394a6361269d1e033949731c85385126dff +size 1370935680 diff --git a/Midm-2.0-Mini-Instruct-Q5_0.gguf b/Midm-2.0-Mini-Instruct-Q5_0.gguf new file mode 100644 index 0000000..811b033 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q5_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:525792fb9cc7f4db362eaa9bdc3282bc7fe31f27f22f54630e13d7f9ff4c76e3 +size 1622561152 diff --git a/Midm-2.0-Mini-Instruct-Q5_K_M.gguf b/Midm-2.0-Mini-Instruct-Q5_K_M.gguf new file mode 100644 index 0000000..280b7ba --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f53ac2a8ef50b8626737919361b417809fd4e334703745d5079d62a64da0a9a8 +size 1654731136 diff --git a/Midm-2.0-Mini-Instruct-Q5_K_S.gguf b/Midm-2.0-Mini-Instruct-Q5_K_S.gguf new file mode 100644 index 0000000..58b19f1 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15153213f84c01aa7f402f5efac79dbf1bba819d6f1ce978871c966c9b680287 +size 1622561152 diff --git a/Midm-2.0-Mini-Instruct-Q6_K.gguf b/Midm-2.0-Mini-Instruct-Q6_K.gguf new file mode 100644 index 0000000..40e23a6 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6e86b6e1bc8464c07f41217cfb4177f22172668f2f75374d2cbec6b9863b365 +size 1897468288 diff --git a/Midm-2.0-Mini-Instruct-Q8_0.gguf b/Midm-2.0-Mini-Instruct-Q8_0.gguf new file mode 100644 index 0000000..876d9a1 --- /dev/null +++ b/Midm-2.0-Mini-Instruct-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:00774ed1a314af5dcc034dfbfea88c9d0e941236c5282b5a1731b0793bf64f88 +size 2455793664 diff --git a/README.md b/README.md new file mode 100644 index 0000000..2fa4475 --- /dev/null +++ b/README.md @@ -0,0 +1,574 @@ +--- +license: mit +language: +- en +- ko +tags: +- KT +- K-intelligence +- Mi:dm +pipeline_tag: text-generation +base_model: +- K-intelligence/Midm-2.0-Mini-Instruct +--- + +

+ + + +

+ +

+🤗 Mi:dm 2.0 Models | +📜 Mi:dm 2.0 Technical Report | +📕 Mi:dm 2.0 Technical Blog* +

+ +

*To be released soon

+ +
+ +## News 📢 + +- 🔜 _(Coming Soon!) GGUF format model files will be available soon for easier local deployment._ +- ⚡️`2025/07/04`: Released Mi:dm 2.0 Model collection on Hugging Face🤗. +
+
+# Table of Contents + +- ___Overview___ + - [Mi:dm 2.0](#midm-20) + - [Quickstart](#quickstart) + - [Evaluation](#evaluation) +- ___Usage___ + - [Run on Friendly.AI](#run-on-friendliai) + - [Run on Your Local Machine](#run-on-your-local-machine) + - [Deployment](#deployment) + - [Tutorials](#tutorials) +- ___More Information___ + - [Limitation](#limitation) + - [License](#license) + - [Contact](#contact) + +
+
+ +# Overview + +### Mi:dm 2.0 + +**Mi:dm 2.0** is a __"Korea-centric AI"__ model developed using KT's proprietary technology. The term __"Korea-centric AI"__ refers to a model that deeply internalizes the unique values, cognitive frameworks, and commonsense reasoning inherent to Korean society. It goes beyond simply processing or generating Korean text—it reflects a deeper understanding of the socio-cultural norms and values that define Korean society. + +Mi:dm 2.0 is released in two versions: + +- **Mi:dm 2.0 Base** + An 11.5B parameter dense model designed to balance model size and performance. + It extends an 8B-scale model by applying the Depth-up Scaling (DuS) method, making it suitable for real-world applications that require both performance and versatility. + +- **Mi:dm 2.0 Mini** + A lightweight 2.3B parameter dense model optimized for on-device environments and systems with limited GPU resources. + It was derived from the Base model through pruning and distillation to enable compact deployment. + + +> [!Note] +> Neither the pre-training nor the post-training data includes KT users' data. + + +
+ +### Quickstart + +Here is the code snippet to run conversational inference with the model: + +```python +import torch +from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig + +model_name = "K-intelligence/Midm-2.0-Mini-Instruct" + +model = AutoModelForCausalLM.from_pretrained( + model_name, + torch_dtype=torch.bfloat16, + trust_remote_code=True, + device_map="auto" +) +tokenizer = AutoTokenizer.from_pretrained(model_name) +generation_config = GenerationConfig.from_pretrained(model_name) + +prompt = "KT에 대해 소개해줘" + +# message for inference +messages = [ + {"role": "system", + "content": "Mi:dm(믿:음)은 KT에서 개발한 AI 기반 어시스턴트이다."}, + {"role": "user", "content": prompt} +] + +input_ids = tokenizer.apply_chat_template( + messages, + tokenize=True, + add_generation_prompt=True, + return_tensors="pt" +) + +output = model.generate( + input_ids.to("cuda"), + generation_config=generation_config, + eos_token_id=tokenizer.eos_token_id, + max_new_tokens=128, + do_sample=False, +) +print(tokenizer.decode(output[0])) +``` + +> [!NOTE] +> The `transformers` library should be version `4.45.0` or higher. + +
+
+ +# Evaluation + + +#### Korean + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
ModelSociety & CultureGeneral KnowledgeInstruction Following
K-Refer*K-Refer-Hard*Ko-Sovereign*HAERAEAvg.KMMLUKo-Sovereign*Avg.Ko-IFEvalKo-MTBenchAvg.
Qwen3-4B53.642.935.850.645.750.642.546.575.963.069.4
Exaone-3.5-2.4B-inst64.067.144.461.359.243.542.443.065.474.068.9
Mi:dm 2.0-Mini-inst66.461.436.770.858.845.142.443.873.374.073.6
Qwen3-14B72.465.749.868.464.155.454.755.183.67177.3
Llama-3.1-8B-inst43.236.433.849.540.733.036.734.860.15758.5
Exaone-3.5-7.8B-inst71.669.346.972.965.252.645.649.169.179.674.4
Mi:dm 2.0-Base-inst89.686.456.381.578.457.358.057.78289.785.9
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
ModelComprehensionReasoning
K-Prag*K-Refer-Hard*Ko-BestKo-Sovereign*Avg.Ko-WinograndeKo-BestLogicKorHRM8KAvg.
Qwen3-4B73.956.791.543.566.667.569.25.656.743.8
Exaone-3.5-2.4B-inst68.758.587.238.062.560.364.17.438.536.7
Mi:dm 2.0-Mini-inst69.555.480.542.561.961.764.57.739.937.4
Qwen3-14B86.774.093.952.076.877.275.46.464.548.8
Llama-3.1-8B-inst59.948.677.431.551.540.126.02.430.919.8
Exaone-3.5-7.8B-inst73.561.992.044.067.264.660.38.649.739.5
Mi:dm 2.0-Base-inst86.570.895.253.076.175.173.08.652.944.8
+ +`*` indicates KT proprietary evaluation resources. + +
+ + +#### English + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
ModelInstructionReasoningMathCodingGeneral Knowledge
IFEvalBBHGPQAMuSRAvg.GSM8KMBPP+MMLU-proMMLUAvg.
Qwen3-4B79.779.039.858.559.190.462.4-73.373.3
Exaone-3.5-2.4B-inst81.146.428.149.741.482.559.8-59.559.5
Mi:dm 2.0-Mini-inst73.644.526.651.740.983.160.9-56.556.5
 
Qwen3-14B83.983.449.857.763.688.073.470.582.776.6
Llama-3.1-8B-inst79.960.321.650.344.181.281.847.670.759.2
Exaone-3.5-7.8B-inst83.650.133.151.244.881.179.440.769.054.8
Mi:dm 2.0-Base-inst84.077.733.551.954.491.677.553.373.763.5
+ +
+ +# Usage + +### Run on Friendli.AI +You can try our model immediately via `Friendli.AI`. Simply click `Deploy` and then `Friendli Endpoints`. + +> [!Note] +> Please note that a login to `Friendli.AI` is required after your fifth chat interaction. + +

+ Left Image + Right Image +

+ +### Run on Your Local Machine +We provide a detailed description about running Mi:dm 2.0 on your local machine using llama.cpp, LM Studio, and Ollama. Please check our [github](https://github.com/K-intelligence-Midm/Midm-2.0) for more information + + +### Deployment + +To serve Mi:dm 2.0 using [vLLM](https://github.com/vllm-project/vllm)(`>=0.8.0`) with an OpenAI-compatible API: +```bash +vllm serve K-intelligence/Midm-2.0-Mini-Instruct +``` + + +### Tutorials +To help our end-users easily use Mi:dm 2.0, we have provided comprehensive tutorials on [github](https://github.com/K-intelligence-Midm/Midm-2.0). +
+ +
+
+ +# More Information + +### Limitation +* The training data for both Mi:dm 2.0 models consists primarily of English and Korean. Understanding and generation in other languages are not guaranteed. + +* The model is not guaranteed to provide reliable advice in fields that require professional expertise, such as law, medicine, or finance. + +* Researchers have made efforts to exclude unethical content from the training data — such as profanity, slurs, bias, and discriminatory language. However, despite these efforts, the model may still produce inappropriate expressions or factual inaccuracies. + + +### License + +Mi:dm 2.0 is licensed under the [MIT License](./LICENSE). + + +### Contact +Mi:dm 2.0 Technical Inquiries: midm-llm@kt.com + +
\ No newline at end of file